Sunday, March 18, 2018

How Science is done?


One of the things that makes humans so special is probably their curiosity, the urge to know, to observe phenomena, to pose questions and above all trying to answer them. Science is one of the ways devised to answer these questions in a systematic and rational manner.

To begin with, I will introduce scientific method. Everything that scientists do can be roughly formulated into an algorithm called the scientific method. I will try to make it more clear using an example.

The first step is to “observe” some phenomena. Observation requires alert mind and constant vigilance. Observation basically includes anything that answers a question “What?”
For instance, we observe that everyday sun is rising in the east (maybe for 20 years of your life and you also know it from historical reports say of at least a few hundred years).

To make sense of it, one then makes a hypothesis, which is basically an answer that may explain the phenomena. This is an important step. It is essential for the hypothesis to be “falsifiable”: meaning that one should be able to disprove the converse of the statement. For instance, the statement that “Sun daily rises in the East” could be a valid hypothesis. Understand that one cannot prove this statement. To prove the statement the observer has to observe the Sun “daily” which is impossible for a normal human being. What can be done instead is to disprove the converse: “Sun doesn't rise in the West”. As far as we don’t see Sun rising in the West provides us with evidence to support “Sun daily rises in the East”. Stronger and increasing shreds of evidence, convert a hypothesis into a fact, a theory. However, a single valid observation of Sun rising in the West is enough to disprove the statement. Generalizing it, one can clearly see that there is nothing called a proof in Science (except maybe in Mathematics where you can provide definitive proofs).

Based on the hypothesis, predictions are made which can then be tested using experiments or analysis or a model can be built that suits well with the data. Experiments are then repeated and reformulated. New predictions are made which are further tested. All of these provide pieces of evidence supporting our hypothesis, never proves it. A single experiment giving a negative result simply crushes the hypothesis or causes it to be restated.

A well-supported hypothesis becomes a theory, a fact. This gives rise to new observations and the cycle continues.

The cycle appears quite simple. However, it does not really take into account all the aspects of “How Science is done?” It does not provide an answer to “What does a Scientist do?” Science like other disciplines requires a lot of hard work and patience. Apart from these steps (mentioned above), a scientist also plays with various parameters. Many experiments fail not because the hypothesis was false but because of errors in instrumentation, in handling and many times, the sources of these errors are difficult to find out. The results appear simple and interesting, but the work done before arriving at it is mostly unknown. The frustration, the excitement, the competition among labs to arrive at the answer first, the politics behind are hardly even recognized.

No comments:

Post a Comment

Statistics and P value

The How Science is done? (previous article) mentions the Scientific Method . One of the most important steps of this algorithm is formula...